Voice leading describes how chord progressions are realised as collections of musical pitches, and how these pitches connect to form simultaneous melodies. The voicer package implements a computational cognitive model of voice leading that has two main applications:
- Quantifying the importance of different aesthetic phenomena (e.g. dissonance, parallel octaves) for determining voice-leading practice in a given musical corpus;
- Generating voice leadings for novel chord progressions.
This model implements various features derived from Huron’s (2001, 2016) perceptual accounts of voice leading. Huron’s principles are grounded in the perceptual phenomenon of auditory scene analysis; the general nature of Huron’s theory is useful for helping the model to generalise between musical styles.